[PATCH] dtc: Remove spurious output on stderr

Simon Glass sjg at chromium.org
Thu Apr 12 09:17:03 EST 2012


Hi David,

On Wed, Apr 11, 2012 at 3:45 PM, David Gibson
<david at gibson.dropbear.id.au> wrote:
> On Wed, Apr 11, 2012 at 11:55:16AM -0700, Simon Glass wrote:
>> Outputing to stderr is best avoided unless there is an error or warning to
>> display. At present dtc always displays the name of the file it is compiling
>> and the input/output formats. For example:
>>
>> DTC: dts->dts  on file "-"
>>
>> This can cause problems in some build systems. For example, U-Boot shows
>> build errors for any boards which use dtc at present. It is typically the
>> only message output during such a build. The C compiler does not output
>> anything in general. The current dtc behaviour makes it difficult to
>> provide a silent build in the normal case where nothing went wrong.
>>
>> The -q flag is currently used to ignore warnings, and this message is not
>> really a warning. If we inserted another level of quietness that just
>> supresses non-warnings, then this would break the current behaviour of -q.
>>
>> Therefore a new flag seems appropriate. Unfortunately both -v and -V are
>> already used, so I have come up with the random choice of -a: "announce
>> operation".
>>
>> This also changes current behaviour, but hopefully in a good way. The main
>> problem is that people will wonder whether dtc actually ran at all, since
>> they are used to seeing the message. A quick check should confirm this, or
>> the -a flag can be added if desired.
>>
>> I'm hoping someone has a better solution.
>
> Ugh.  Don't bother with the option, just remove the message.  I've
> been half meaning to get rid of it for ages.

:-) That's not the answer I expected. Will do.

>
> --
> David Gibson                    | I'll have my music baroque, and my code
> david AT gibson.dropbear.id.au  | minimalist, thank you.  NOT _the_ _other_
>                                | _way_ _around_!
> http://www.ozlabs.org/~dgibson

Regards
Simon


More information about the devicetree-discuss mailing list